Next Article in Journal
Identification of Potential Therapeutic Targets for Coronary Atherosclerosis from an Inflammatory Perspective Through Integrated Proteomics and Single-Cell Omics
Previous Article in Journal
Molecular and Clinical Aspects of Osteogenesis Imperfecta Type VI: A Case Series with Novel SERPINF1 Gene Variants
Previous Article in Special Issue
A New Adjuvant Treatment for Glioblastoma Using Aprepitant, Vortioxetine, Roflumilast and Olanzapine: The AVRO Regimen
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJMS
Volume 26
Issue 13
10.3390/ijms26136198
ijms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Exploring the Role of Peripheral Macrophages in Glioma Progression: The Metabolic Significance of Cyclooxygenase-2 (COX-2)

by
Jens Pietzsch
1,2,
Magali Toussaint
3,
Cornelius Kurt Donat
4,
Alina Doctor
1,2,
Sebastian Meister
1,
Johanna Wodtke
1,
Markus Laube
1,
Frank Hofheinz
5,
Jan Rix
6,
Winnie Deuther-Conrad
3 and
Cathleen Haase-Kohn
1,*
1
Department of Radiopharmaceutical and Chemical Biology, Institute of Radiopharmaceutical Cancer Research, Helmholtz-Zentrum Dresden-Rossendorf (HZDR), 01328 Dresden, Germany
2
School of Science, Faculty of Chemistry and Food Chemistry, Technische Universität Dresden, 01062 Dresden, Germany
3
Department of Neuroradiopharmaceuticals, Institute of Radiopharmaceutical Cancer Research, Helmholtz-Zentrum Dresden-Rossendorf, 04318 Leipzig, Germany
4
Department of Medicinal Radiochemistry, Institute of Radiopharmaceutical Cancer Research, Helmholtz-Zentrum Dresden-Rossendorf, 01328 Dresden, Germany
5
Department of Positron Emission Tomography, Institute of Radiopharmaceutical Cancer Research, Helmholtz-Zentrum Dresden-Rossendorf, 01328 Dresden, Germany
6
Department of Medical Physics and Biomedical Engineering, Faculty of Medicine Carl Gustav Carus, TU Dresden, Fetscherstrasse 74, D-01307 Dresden, Germany
*
Author to whom correspondence should be addressed.
Int. J. Mol. Sci. 2025, 26(13), 6198; https://doi.org/10.3390/ijms26136198
Submission received: 17 March 2025 / Revised: 18 June 2025 / Accepted: 21 June 2025 / Published: 27 June 2025
(This article belongs to the Special Issue Molecular and Cellular Research on the Glioma and Tumor Microenvironment)
Browse Figures
Review Reports Versions Notes

Abstract

Glioblastoma (GBM) is the most aggressive form of malignant gliomas, with the eicosanoid-synthesizing enzyme cyclooxygenase-2 (COX-2) playing a pivotal role in its progression via the COX-2/prostaglandin E2/4 axis. COX-2 upregulations in tumor cells induces a pro-inflammatory tumor microenvironment (TME), affecting the behavior of invading bone marrow-derived macrophages (Mϕ) and brain-resident microglia (MG) through unclear autocrine and paracrine mechanisms. Using CRISPR/Cas9 technology, we generated COX-2 knockout U87 glioblastoma cells. In spheroids and in vivo xenografts, this resulted in a significant inhibition of tumorigenic properties, while not observed in standard adherent monolayer culture. Here, the knockout induced a G1 cell cycle arrest in adherent cells, accompanied by increased ROS, mitochondrial activity, and cytochrome c-mediated apoptosis. In spheroids and xenograft models, COX-2 knockout led to notable growth delays and increased cell death, characterized by features of both apoptosis and autophagy. Interestingly, these effects were partially reversed in subcutaneous xenografts after co-culture with Mϕ, while co-culture with MG enhanced the growth-suppressive effects. In an orthotopic model, COX-2 knockout tumors displayed reduced proliferation (fewer Ki-67 positive cells), increased numbers of GFAP-positive astrocytes, and signs of membrane blebbing. These findings highlight the potential of COX-2 knockout and suppression as a therapeutic strategy in GBM, particularly when combined with suppression of infiltrating macrophages and stabilization of resident microglia populations to enhance anti-tumor effects.

Graphical Abstract

1. Introduction

Glioblastoma (GBM) is the most aggressive primary brain cancer, with a median survival of only 8 to 18 months, despite typically agressive surgical resection, chemo- and radiotherapy [1,2]. A crucial factor contributing to GBM progression and invasiveness is the tumor microenvironment (TME), a dynamic ecosystem of both neoplastic and non-neoplastic cells [3,4]. Among these, tumor-associated macrophages (TAMs) are par-ticularly prominent, comprising 40–50% of the tumor mass and correlating with higher tumor grade and shorter patient survival [5]. TAMs encompass brain-resident microglia (MG) and bone marrow-derived macrophages Mϕ, both playing pivotal roles in GBM pathogenesis [4].
A key driver within the TME are different inflammatory processes, heavily mediated by TAMs [6]. At the molecular level, the Cyclooxygenase-2 (COX-2)-prostaglandin E2/E4 (PGE2/PGE4) axis emerges as a critical pathway [6,7,8]. COX-2 expression is induced in TAMs and constitutively expressed in human adherent GBM cell lines, correlating with tumor grade and poor prognosis [9,10,11]. Interestingly, this pathway is predominantly lo-calized in hypoxic areas, where hypoxia-inducible factor-1α (HIF-1α) further enhances COX-2 expression [11,12,13]. Hypoxia not only seems to drive COX-2 upregulation, but also modulates TAM-tumor cell communication through altered ligand-receptor interactions, thereby promoting tumor progression. Additionally, hypoxia and associated oxidative stress amplify inflammation, reactive oxygen species (ROS) production, and apoptosis, further increasing tumor aggressiveness [14,15,16].
Given the critical role of the COX-2/PGE2 axis in GBM, therapeutic targeting of this pathway offers potential benefits, including reduced cell proliferation, angiogenesis, and enhanced apoptosis [17,18]. However, translating these findings into clinical strategies requires appropriate models to dissect the complexity of TAM-GBM interactions.
In this study, we aim to further elucidate the role of COX-2 in the TAM-GBM dynamic TEM. Employing genetically modified human glioblastoma U87 cells with a complete COX-2 knockout (U87COX-2KO), we investigated metabolic interactions with TAM popula-tions, including bone marrow-derived macrophages and microglia. Comparisons were made with U87 wild-type cells (U87wt) in adherent cultures, spheroids, and corresponding heterotopic xenograft mouse models. Finally, to better evaluate the influence of the COX-2/PGE2 axis on direct TAM-tumor interactions and tumor growth, an appropriate orthotopic xenotransplantation model was employed.

2. Results

2.1. In Vitro Experiments in Adherent Cell Culture

The cell growth rate of U87wt and U87COX−2KO cells under normoxia and a sufficient supply of nutrients was not statistically different (Figure 1a). The stable COX-2 knockout in U87 cells, confirmed with RFP flow cytometry (Figure S1a–c), resulted in a significant increase in G1 cell cycle arrest under normoxic (* p < 0.05) and hypoxic conditions compared to U87wt cells (Figure 1b; Supplementary Materials: Figure S1d).
Antibody array analysis reveals an upregulation of apoptosis-related proteins (e.g., MAPK15 and p53) and downregulation of tumor–TME crosstalk markers (e.g., Il-10, CXCL2, and E-Selectin) in COX-2 knockout cells (Figure 1c). PGE2 levels were unaffected by COX-2 knockout, suggesting a COX-1-mediated compensatory pathway during arachidonic acid metabolism (Figure S2a,b). Clonogenic survival analysis after X-ray radiation indicated significantly reduced plating efficiency in U87COX−2KO at higher radiation doses (4–10 Gy) (Figure S3).
In monolayer cultures, COX-2 knockout significantly increases the number of meshes and segments (* p < 0.05), as demonstrated by the angiogenesis network formation on the hCMEC endothelial cells (Figure S4). Interestingly, U87COX−2KO shows significantly increased cell migration (up to 72 h, ** p < 0.01) and invasion (up to 84 h, *** p < 0.001) (Figure S5a,b). This trend persisted under nutrient deprivation and after stimulation with activated macrophage supernatants, suggesting an elevated response to inflammatory stimuli (Figure S5c).
Western blot analysis confirms the successful COX-2 knockout in U87COX−2KO cells (Figure 1d). No difference in COX-1 expression was observed between U87wt and U87COX−2KO cells, indicating that COX-2 knockout did not trigger the compensatory upregulation of COX-1. Similarly, monocytes (MCs) and microglia (MG), lacking COX-2 expression, showed no change in COX-1 levels, and reduced expression for MG. Interestingly, lysyl oxidase (LOX), a protein associated with tumor progression, was reduced in U87COX-2KO and MG cells. Additionally, U87COX-2KO and MC cells exhibited significantly higher cytochrome c release from mitochondria, compared to wildtype cells.
COX-2 knockout cells showed significantly reduced basal mitochondrial activity (Figure 1e, upper panel), but this increased under stress conditions, such as nutrient deprivation or inflammatory stimulation.
ROS activity (Figure 1e, middle panel) and malondialdehyde (MDA) (Figure 1e, lower panel) levels were consistently found to be significantly higher in COX-2 knockout cells across all tested conditions, including oxidative stress (H2O2) and hypoxia-inducing treatments (CoCl2).
Under normoxic conditions, U87COX−2KO cells exhibited significantly reduced glucose (*** p < 0.001) (Figure 1f, upper left panel) and significantly increased [18F]FMISO uptake (*** p < 0.001) (Figure 1f, upper right panel). Together, this indicates higher intrinsic hypoxia in comparison to U87 wildtype cells.
When U87wt cells were co-cultured with microglia, a significantly reduced [18F]FMISO radiotracer uptake was observed. All other conditions inducing either oxidative, hypoxic, or nutritional deprivation were not affected by being co-cultured with microglia (Figure 1f, lower panel left). U87COX−2KO cells also responded with reduced [18F]FMISO uptake to being co-cultured with microglia under all stress conditions (***p < 0.001) (Figure 1f, lower right panel). COX-2 knockout cells displayed higher hypoxia under stress compared to U87wt, except for oxidative stress, where no effect was observed. By comparing U87COX−2KO with U87wt, a significantly increased [18F]FMISO uptake was observed for COX-2 knockout cells after treatment with either S_HMC3, woFCS, or woGlucose (### p < 0.001) and with CoCl2 (# p < 0.05). In contrast, COX-2 knockout cells after co-culture with MG displayed a significant decrease in [18F]FMISO uptake after treatment with various compounds compared to U87wt + MG (§ p < 0.05 for CoCl2, S_HMC3, and woGlucose; §§ p < 0.01 for H2O2).

2.2. In Vitro Experiments in Spheroids

COX-2 knockout spheroids exhibited significantly reduced growth compared to wildtype spheroids, stalling between days 5 and 19 (Figure 2a).
Spheroids from knockout cells showed a lower core cell density, as confirmed by H&E (Figure 2b) staining, and altered mechanical properties, revealed by Brillouin spectroscopy (Figure S6). Hypoxia regions, identified using pimonidazole staining, were more pronounced in COX-2 knockout spheroids, aligning with decreased Ki-67-positive cells (Figure 2b).
Spheroid glucose uptake ([18F]FDG) was significantly reduced in U87COX−2KO compared to U87wt, correlating with lower viability and slower growth (*** p < 0.001) (Figure 2c). However, the hypoxia tracer uptake ([18F]FMISO) did not differ significantly between the spheroids from both cell types (Figure 2c). ROS levels in U87COX−2KO spheroids significantly increased over time (### p < 0.001), contrasting with the decline observed in U87wt spheroids (Figure 2d).
In U87COX−2KO spheroids, a decreased expression of COX-1, the endothelial cell adhesion molecule E-selectin, the cytokine CXCL2, Fibronectin, and the monocyte/macrophage marker CD68 was found compared to U87wt spheroids. CD31, an endothelial cell marker, remained unchanged (Figure 2e). We observed a slight increase in CD44 expression in the U87wt spheroids over time. In contrast, the COX-2 knockout spheroids showed a slightly higher CD44 expression on day 5 and 12, which then was lower on day 19 compared to the wild-type spheroids. (Supplementary Materials: Figure S10a).

2.3. In Vivo Effects of COX-2 Knockout in Subcutaneous Xenograft Models

The growth rates of U87COX−2KO tumors, compared to U87wt, were significantly slower in a subcutaneous xenograft model (*** p < 0.001) (Table 1 and Table 2; Figure 3a). The relative growth rate (change of tumor size in percent of the initial size per week), calculated from the reciprocal value of the doubling rate (days), of U87COX−2KO was only 50% (0.9 ± 0.4) of U87wt (2.0 ± 1.0).
Two distinctive patterns were observed when U87COX−2KO cells were co-cultured with either bone marrow-derived MCs/Mϕs or MG cells prior to injection. The suppression of the growth rate was reversed when COX-2 knockout cells were co-cultured with bone marrow-derived MC (2.0 ± 0.7) or Mϕ (1.8 ± 0.7) cells (*** p < 0.001) (Figure 3b, right panel), This suggests a critical role for infiltrating immune cells in modulating tumor growth compared to U87wt co-cultured with MCs or Mϕs (Figure 3b, left panel). In contrast, co-cultivation with microglia (MG) (0.4 ± 0.6) further enhanced the suppression of tumor growth in COX-2 knockout cells (** p < 0.01) (Figure 3c, right panel) compared to U87wt cells (Figure 3c, left panel).
Subcutaneous COX-2 knockout tumors and multi-cellular tumors with MG displayed significantly reduced [18F]FMISO uptake compared to U87wt and U87wt + MG, respectively (*** p < 0.001) (Figure 3d, left panel). However, no difference in glucose ([18F]FDG) uptake was observed between the two groups (Figure 3d, right panel).

2.4. In Vivo Effects of COX-2 Knockout in Orthotopic Xenograft Models

Orthotopic xenografts confirmed prior findings of the subcutaneous xenograft model. COX-2 knockout tumors exhibited a significantly reduced volume compared to those from wildtype cells (* p < 0.05). In eight out of eight animals injected with U87wt, the average tumor volume was 3.72 mm3 at 27.75 days (Figure 4a). In five out of eight animals injected with U87COX−2KO, the average tumor volume was 1.51 mm3 at 67.2 days (Figure 4b).
H&E staining of U87COX−2KO tumors revealed altered cellular morphology, with large, bubbled cells indicative of increased necrosis or apoptosis, which was not observed in U87wt (Figure 5a).
Quantitative histology using HALO® confirmed significantly reduced Ki-67-positive cells per mm2 in U87COX−2KO tumors (** p < 0.01) (Figure 5b), despite no difference in overall cell density (Figure S7b). Analysis of Ki-67 staining intensity (weak, moderate, and strong, based on identical thresholds across all sections) showed a significant genotype effect [F(1,24) = 45.47, p  <  0.0001], with reductions seen across all intensity levels. A post hoc test revealed that reductions in KI-67-positive cells were equally driven by all cells of U87COX−2KO (p > 0.01 and p < 0.001), irrespective of the staining intensity (Figure S7a).
Immunostaining for microglia (IBA1-positive) showed no difference in density within the tumor (Figure 5c) and an area of 150 µm around it (Figure S7c). Albeit generally low across both tumor entities, the density of GFAP-positive astrocytes was significantly higher within U87COX−2KO tumors. This effect was even more pronounced in the surrounding 150 µm area, where glial scar formation was prominent (p > 0.05; Figure 5d; Figure S7d).
Segmentation validation with isotype controls confirmed negligible false positives for Ki-67 (≤ 2.28 cells/mm²) (Figure S8a) and none for IBA1 (Figure S8b). Percent area-based analysis replicated the findings for KI-67, IBA1, and GFAP (Figure S9a–c), supporting the accuracy of our segmentation-based quantitative approach. A tendency towards decreased CD44 expression was found in tissue sections from U87COX-2KO cells, orthotopically implanted cells, when compared sections of animals implanted with wildtype cells (Supplementary Materials: Figure S10b).

3. Discussion

Pathophysiological upregulation of COX-2 is strongly linked to GBM progression and poor prognosis. In contrast, expression in healthy tissue remains low, primarily localized to macrophages and microglia [9]. Within the glioma TME, both tumor cells and TAMs utilize the COX-2/PGE2 axis in autocrine and paracrine signaling [17,18]. This interaction is complex, involving multiple feedback loops, and it is unclear whether immune cells alter tumor growth or are influenced by it [3,19]. Unlike peripheral cancers, where the immune response can be suppressed via pathways like PD1/PD-L1 [20], GBM progression is shaped by the unique brain environment, including the blood–brain barrier and glial cell activity [21]. The enzyme COX-2 significantly impacts cell proliferation, invasion, metastasis, and immunosuppression, making it a promising therapeutic target in GBM [22]. The U87 glioblastoma model was chosen due to its wide application as a GBM model. Furthermore, it allows the interpretation of the obtained results in comparison with those already described for the model [23].
Additional inhibition of COX-2 in the TME was beyond the scope of this study. Instead, we focused on the effect of a complete COX-2 knockout with regard to tumor growth and its effect on the tumor microenvironment and vice versa. The CRISPR/Cas9-mediated COX-2 knockout in U87 glioblastoma cells significantly suppressed growth in spheroids and xenograft models but not in adherent cells. This suggests that COX-2 primarily impacts the interplay between tumor cells and the TME rather than intrinsic tumor cell proliferation. Measurable changes in adherent COX-2 knockout cells indicate G1 cell cycle arrest, increased intrinsic hypoxia, reactive oxygen species (ROS) formation, and mitochondrial activity, triggering or accelerating apoptotic and potentially autophagic pathways.
Previous studies have shown that the inhibition, silencing, or deletion of COX-2 can alter cell cycle progression, proliferation, and invasion in various tumor models [9,24,25,26]. In this study, we demonstrate that U87COX-2KO glioblastoma cells predominantly exhibit G1 cell cycle arrest without affecting in vitro proliferation. This arrest may persist until the cells receive signals to either resume proliferation or undergo cell death, aligning with similar findings in hepato-cellular carcinoma cells after COX-2 silencing [25]. Based on our data we assumed that the G1 arrest does not inevitably lead to apoptosis in vitro. Under optimal nutrient conditions, cells resume proliferation, and the G1 arrest is only transient but detectable. However, in nutrient-limited environments—such as spheroids or subcutaneous/orthotopic models—this balance collapses, resulting in reduced proliferation and induction of apoptosis.
It is important to note that COX-2 expression can vary considerably between cell lines of a tumor entity and, more generally, between cell lines of different tumor entities [27,28]. Influences on COX-2 expression, e.g., through experimental pharmacological interventions, are also characterized by high variability [29]. This makes direct comparisons between studies or generalizable conclusions difficult. As published previously, COX-2 inhibitors are able to reduce proliferation and colony formation in U87MG neurospheres in vitro [30]. However, we observed no reducing effects of COX-2 knockout on proliferation, colony formation, or motility in vitro, consistent with prior findings. We rather assume that off-targets effects of COX-2 inhibitors could be responsible [24,26]. Similarly, COX-2 knockout in A2058 melanoma cells did not affect proliferation but reduced invasion and was found not to release prostaglandin E2 (PGE2) [26]. Adherent U87wt and U87COX-2KO cells exhibited low basal levels of PGE2, which were unaffected by the COX-2 status. However, PGE2 levels significantly increased upon incubation with arachidonic acid. While PGE2 production during inflammation is predominantly mediated by the COX-2 pathway [31], we assume it can also occur via COX-1 and cytosolic prostaglandin E2 synthase in the central nervous system [32,33].
U87COX−2KO cells show higher intrinsic hypoxia, ROS, and mitochondrial activity, which was consistent even under further stressors mimicking the TME. Normoxia and hypoxia in cell cultures differ significantly from tissue oxygen levels, with standard 20% O₂ incubators exceeding the physiological peri-cellular oxygen levels. Accurate hypoxia models should mimic in vivo oxygen levels when known [34]. The elevated [¹⁸F]FMISO uptake in adherent COX-2 knockout cells may result from self-inflicted hypoxia due to high oxygen consumption for ROS production or increased mitochondrial density [35], leading to inhibited growth and enhanced apoptosis in spheroids and in vivo models. It can be assumed that the upregulation of cytochrome c in U87COX−2KO cells is associated with mitochondrial membrane breakdown due to increased ROS or apoptotic conditions involving Bcl-2 [36,37,38]. Cancer signaling pathway arrays showed significant upregulation of MAPK15 and p53, likely driven by elevated ROS, promoting apoptosis and autophagy [39,40]. This aligns with reports of COX-2 inhibitors affecting survival and apoptosis pathways [41].
In spheroids, which better mimic in situ tumors with organized proliferative and necrotic regions, COX-2 knockout significantly impaired growth and proliferation. Similar findings were reported for melanoma cells and prostate cancer, where COX-2 or mPGES-1 inhibition reduced PGE2 production and spheroid growth [26,42]. COX-2 knockout also downregulated proteins involved in tumor–macrophage crosstalk, including CXCL2 and E-selectin, with CXCL2 being a pro-angiogenic chemokine linked to glioma progression and a potential therapeutic target [43,44,45]. The altered metabolism and increased ROS in U87COX−2KO cells may activate anti-tumor pathways via apoptosis, contributing to impaired tumor growth. The discrepancies between the in vitro - and spheroid/in vivo data may be due to COX-2 influencing tumor growth indirectly via the microenvironment rather than directly affecting cancer cell proliferation. Under optimal in vitro conditions, COX-2 knockout effects can likely be compensated. This balance is disrupted under more complex, tumor-like conditions in spheroids and in vivo, revealing the inhibitory impact. In vitro, cells may remain in a survival mode that cannot be sustained in the presence of microenvironmental stress.
Consistent with the spheroid findings, U87COX−2KO cells exhibited suppressed proliferation and tumor growth in both subcutaneous and orthotopic xenograft models. This aligns well with previous melanoma studies [26] and reports on COX-2 knockdown in other cell lines [24,25,46]. To investigate the tumor-associated macrophage (TAM)–glioma cell crosstalk, multi-cellular tumors with human monocytes (MCs), macrophages (Mϕs), or microglia (MG) were implanted. Peripheral monocytes and macrophages tended to accelerate U87wt tumor growth rates. More importantly, their presence abolished the growth suppression observed for U87COX−2KO, while microglia further it. These findings highlight TAMs’ crucial role in glioma development and progression. Similar effects have been observed in vitro, where microglia co-cultures affected glioblastoma cell proliferation and migration, and even induced drug resistance at low ratios [47]. This underscores the complexity of TAM–tumor interactions in driving glioblastoma progression.
Interestingly, CD44, a known GBM stem cell marker [48], showed a trend towards reduced—though not significant—expression in orthotopic xenografts after COX-2 knockout. In COX-2 knockout spheroids, CD44 reduction was only observed on day 19. This, in part, is consistent with data published by Lombardi et al., who found that COX-2 upregulation (via temozolomide) promoted an immunosuppressive microenvironment with increased CD44 expression. In turn, COX-2 inhibition via celecoxib reduced CD44 expression and counteracted this effect, which was reversed by exogenous prostaglandin E2, underscoring the role of COX-2 or respective prostaglandins [18].
In an orthotopic glioblastoma model, a trend towards increased numbers of IBA-positive microglia with amoeboid morphology was observed at tumor margins, consistent with prior findings [49]. However, COX-2 knockout did not affect IBA-positive cell density within the tumor margins or stroma, even though their overall morphology was drastically different from the resting microglia in both U87 entities. In contrast, GFAP-positive astrocytes were moderately increased in and around U87COX−2KO tumors, potentially due to more efficient glial scar formation or autophagy, as the GFAP has been implicated in mitochondrial transfer and autophagy pathways via p38/MAPK and mTOR downregulation [50,51]. COX-2 expression and PGE2 secretion were previously shown to be upregulated in U87 cells co-cultured with M2 macrophages, promoting vascularization through COX-2 activation and highlighting crosstalk between inflammatory and perivascular microenvironments [52]. Similarly, phorbol ester-treated THP-1 cells enhanced cancer cell invasion and angiogenesis via COX-2-dependent MMP-9, VEGF-A, and bFGF release [53]. The factors driving TAM–glioma cell crosstalk remain unclear, but TAMs and tumor cells dynamically alter their expression profiles based on environmental signals [19]. Array data suggest COX-2 knockout downregulates proteins essential for TAM interaction, while the upregulation of p53 and MAPK15 may promote apoptosis and autophagy in knockout cells [54].

4. Materials and Methods

4.1. Cell Culture Conditions

The human glioblastoma cell line U87 MG (# HTB-14) and human microglia cell line HMC3 (CRL-3304; referred to as MG throughout the text) were purchased from ATCC. The human acute monocytic leukemia cell line THP-1 (ACC-16) was purchased from DSMZ. THP-1 monocytes (MCs) were differentiated into M0 macrophages (Mϕs) using 64 nM PMA (Phorbol 12-myristate 13-acetate) for three days.
All cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) fetal calf serum and 1 U/mL penicillin/streptomycin (all reagents from Biochrom, Berlin, Germany) under normoxic conditions (37 °C, 5% CO2). For hypoxia experiments (extrinsic hypoxia model), cells were cultured at a reduced oxygen level (<1% v/v) in a special incubator equipped with an oxygen sensor (Gasboy; Labotect, Göttingen, Germany). To generate spheroids, U87wt and U87COX−2KO cells were harvested by trypsinization and seeded in 96-well non-adherent plates (Greiner Bio One, Frickenhausen, Germany) at a density of 2000 cells per well [in 100 μL of a 0.5% methylcellulose (Bio-Techne, Wiesbaden, Germany) and culture medium]. Spheroids were maintained for 5, 12, and 19 days.

4.2. Knockdown of COX-2 with CRISPR/Cas9

A CRISPR/Cas9 knockdown kit targeting human COX-2 was obtained from Santa Cruz Biotechnology (Dallas, TX, USA). U87 cells (2 × 105) were seeded into six-well plates 24 h prior to transfection. Cells were transfected with 1 µg COX-2 CRISPR/Cas9 KO Plasmid and 1 µg COX-2 HDR Plasmid using UltraCruz® Transfection Reagent (Dallas, TX, USA) in a serum-free growth medium, following the manufacturer’s protocol.
After 24 h, the medium was replaced with a complete growth medium containing 0.5 µg/mL puromycin for selection. Transfected cells were monitored via RFP fluorescence microscopy, and RFP-positive and puromycin-resistant cells were collected for monoclonal expansion. COX-2 knockout was confirmed by fluorescent microscopy and Western blotting.

4.3. Antibody Array

Cancer Biomarker (247 targets) and Cancer Signaling Phospho (269 targets) antibody arrays (tebu-bio, Offenbach, Germany) were used following the manufacturer’s protocol. Array images were assessed using tebu-bio. Mean values and statistical analysis was provided by the manufacturer. Changes were considered significant if the expression ratio between U87wt and U87COX-2KO was ≤ 0.5 (decrease) or ≥ 2.0 (increase). A value of 0.5 indicates that the protein amount has decreased by 50%, and a value of 2 means the protein amount has doubled.

4.4. Growth Rate Analysis

Cell growth was assessed using the IncuCyte® Live-Cell Analysis System (Sartorius, Göttingen, Germany). Cells were seeded in 96-well plates at ~10% confluence and imaged every 2 h for 9 days. Growth rates were determined using the system’s integrated confluence algorithm.

4.5. RFP Control and Cell Cycle Analysis by FACS

RFP fluorescence in U87COX−2KO was verified before each experiment using an Attune NxT CytKick flow cytometer (Thermo Fisher Scientific, Dreieich, Germany). For cell cycle analysis, ethanol-fixed cells were treated with 50 µg/mL RNase A and stained with 10 µg/mL propidium iodide before analysis by flow cytometry. The results were analyzed by FlowJo 7.6.1 software (FlowJo, LLC, Ashland, OR, USA).

4.6. Western Blot Analysis

SDS-PAGE and Western blotting were performed, as described previously [16]. PVDF membranes were incubated with anti-COX-2 (D5H5) XP® (#12282; 1:400, Cell Signaling, Danvers, MA, USA); anti-COX1 (ab227513; 1:500, Abcam, UK); Cytochrome c (#MA5-15078; 1:400; Invitrogen, USA); anti-Phospho-MAPK15 (#PA5-99135, 1:500, Invitrogen, USA); anti-LOX (F-8) (sc-373995, 1:500, Santa Cruz Biotechnology, INC.; Dallas, Tx, USA); anti-GAPDH (G8795, 1:1000; Sigma-Aldrich, Darmstadt, Germany); and anti-β-actin (A5060; 1:1000, Sigma-Aldrich, Germany) antibodies, followed by incubation with an appropriate peroxidase-coupled secondary antibody (anti-rabbit IgG, A0545, Sigma-Aldrich, 1:5000; anti-mouse IgG, A9044, Sigma-Aldrich, 1:10.000). Bands were visualized using SuperSignal® West Pico and Femto chemiluminescent Substrate (Thermo Fisher Scientific, Dreieich, Germany) and imaged with an MF-ChemiBIS Bio-Imaging System (Biostep GmbH, Burkhardtsdorf, Germany). Western blot densitometry and relative intensity of protein bands calculated using Aida 5.10 software.

4.7. Histological Analysis of U87wt and U87COX−2KO Spheroids

On days 5, 12, and 19, the spheroids were fixed with phosphate-buffered paraformaldehyde (4%) for 30 min and processed in an alcohol series ending with isopropanol, followed by paraffin embedding. For the histological examination, 2 µm thick sections were mounted on SuperFrost Plus slides. Sections were deparaffinized, rehydrated, and stained with hematoxylin and eosin (H&E).

4.8. Spheroid Histology and Immunohistochemistry

The spheroids were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned (2–5 µm). Hematoxylin and eosin (H&E) staining was performed for structural analysis. For hypoxia detection, the spheroids were treated with 200 µM pimonidazole for 2 h before fixation, and subsequent immunohistochemical staining was performed for Ki-67 (Abcam, Cambridge, UK, #15580, 1:2000), CXCL2 (GRO-β) (Bio-Rad, Dreieich, Germany, #150439, 1:500), CD31 (Abcam #28346, 1:75), CD68 (Abcam, #125212, 1:500), Fibronectin (Abcam, #2413, 1:250), CD62E (E-selectin) (Abcam #18981, 1:500), and COX-1 (Abcam #109025, 1/:50). Stained slides were imaged using an AxioImager A1 microscope (Zeiss, Jena, Germany).

4.9. Oxidative Stress and Metabolic Assay

Reactive oxygen species (ROS) were measured using CellROX® DeepRed (Thermo Fisher Scientific). Mitochondrial activity was assessed using MitoTracker® Green FM dye. Lipid peroxidation was quantified by a thiobarbituric acid reactive substances (TBARS) assay (Cell Biolabs, San Diego, CA, USA).
For this purpose, U87wt and U87COX−2KO cells were treated with different agents for 24 h: H2O2 (200 μM) as a model of an oxidative stressor; CoCl2 (200 nM) as a hypoxia-inducing agent; supernatants of HMC3 microglia cells (S_HMC3); media without serum (woFCS); and media without glucose (woGlucose).

4.10. Radiotracer Uptake Assay

Glucose ([18F]FDG) and hypoxia ([18F]FMISO) were assessed in cells and spheroids. Cells were incubated with 0.5 MBq [18F]FDG for 1 h or 0.5 MBq [18F]FMISO for 4 h, washed, lysed, and radioactivity measured using a gamma counter. The protein-normalized uptake was reported as the percent initial dose (ID) per mg protein.

4.11. Subcutaneous Glioblastoma Xenografts

Animal experiments were conducted in accordance with German Animal Welfare Regulations and approved by the Landesdirektion Dresden Animal Ethics Committee (reference number DD24.1-5131/449/49). Female NMRI nude mice (Janvier Labs) were subcutaneously injected with 5 × 10⁶ U87wt, U87COX−2KO, or Mϕ cells, or 1 × 10⁶ MC or MG cells, each suspended in 100 µL PBS/Matrigel (50:50). Tumor size was measured three times per week, and volume was calculated as V = π/6 × (tumor length × tumor width2). The growth rate and doubling time were determined from exponential curve fitting. Non-growing tumors were assigned a growth rate and doubling time of zero at the time of the last measurement. Statistical comparisons were performed using the Mann–Whitney U test (p < 0.05) performed with R (v4.3.1).

4.12. Small Animal PET/CT Imaging of Subcutaneous Glioblastoma Xenografts

Small animal positron emission tomography (PET) was performed using a nanoPET/CT scanner (Mediso Medical Imaging Systems, Budapest, Hungary) with a field of view (FOV) of 9.6 × 10 cm, enabling the whole-body imaging of mice. U87wt and U87COX−2KO tumor-bearing mice were anesthetized with 9% desflurane (Suprane, Baxter, Heidelberg, Germany) in 30% oxygen/air (1 L/min) and intravenously injected with 8–12 MBq (228–342 MBq/kg) of [¹⁸F]FMISO or [¹⁸F]FDG in 0.2 mL of saline via a tail vein catheter. Imaging was performed dynamically for 60 min ([¹⁸F]FDG) or 30 min ([¹⁸F]FMISO, 240 min post-injection).
CT scans were recorded for anatomical referencing and attenuation correction. Data were reconstructed using the Tera-Tomo™ 3D algorithm and analyzed with ROVER software v3.0.73h (ABX, Radeberg, Germany), generating maximum intensity projections (MIPs). Tumors were delineated to create 3D volumes of interest (VOIs). Standardized uptake values (SUVmean and SUVmax) were calculated for VOIs and analyzed using GraphPrism9 Software, San Diego, CA, USA).

4.13. Orthotopic Glioblastoma Xenograft Tumor Model

Female athymic nude mice (Rj:NMRI-Foxn1 nu/nu) were used for tumor implantation at the age of 8 weeks (25–30 g), as outlined in Figure 6b. During microsurgery, mice were anesthetized with a mixture of oxygen/air and desflurane (9–12% depending on respiratory rates) under aseptic conditions and gentle heating. Animals were placed into a Stoelting stereotactic frame (just for mouseTM, Stoelting Europe, Dublin, Ireland). A midline incision was performed, and a burr hole drilled, 0.5 mm anterior and 2.4 mm lateral to the bregma. About 1.6–8 × 104 U87wt or U87COX−2KO cells, respectively, were suspended in Hank’s Buffered Salt Solution (HBSS, 1×). This suspension was injected at 3.0 mm depth into the brain parenchyma with a flow of 0.1 μL/min using a 10 μL Hamilton syringe (Figure 6a). After injection, the burr hole was closed with bonewax (Ethicon, Cincinnati, OH, USA), the scalp incision was sutured (Vicryl 6.0, Ethicon), and the surface was antiseptically cleaned.

4.14. MRI Imaging

Tumor volumes were measured using a dedicated 7T small animal magnetic resonance imaging scanner (MRI, ParaVision software 6.0.1., Bruker, Ettlingen, Germany), as outlined in Figure 6b. For MRI measurements, a T2-weighted measuring sequence (TRARE) was applied with an echo and repetition time of 35 ms and 2500 ms, respectively. The spatial resolution was set to 78 µm in an xy direction. The slice thickness was set to 0.7 mm. Hereinafter, tumor volumes were quantified using the software ROVER (version v3.0.73h; ABX GmbH, Radeberg, Germany). To determine the tumor size, all voxels of the tumor in the slices were manually selected, and the volume of the generated ROI was calculated.

4.15. Immunohistochemistry of U87wt and U87COX−2KO Brain Tumors

The experimental pipeline of tissue cutting, staining, and analysis is shown in Figure 6c,d. Brain tumors were collected from mice with MRI-confirmed growth near the endpoints. After deep anesthesia and cervical dislocation, animals were perfused, and their brains were extracted. They were further fixed in 4% PFA for 24 h and stored in PBS with 0.1% sodium azide and 20% sucrose for 3 days. The brains were paraffin-embedded, and 5 µm coronal sections were prepared, with 3–4 sections per slide and a spacing of 75–500 µm using a microtome (Microm HM340E; Thermo Fisher Scientific, Cambridge, UK).
The sections were first deparaffinized and rehydrated. For hematoxylin and eosin (H&E) staining [27], the sections were immersed in Mayer’s hematoxylin for three minutes, followed by blueing, counterstaining with eosin for 30 s, and dehydration through an ethanol gradient. Finally, slides were cleared with RotiClear® and mounted using RotiMount® (Carl Roth, Karlsruhe, Germany) For immunohistochemistry, antigen retrieval was performed by heating sections in a 10 mM citrate buffer at 95 °C for 20 min. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide, and nonspecific binding was blocked with 10% fetal calf serum in TBS–Tween for one hour at room temperature. The sections were incubated overnight at 4 °C with primary antibodies targeting the glial fibrillary acidic protein (GFAP, Agilent, Santa Clara, CA, USA) #Z0334, 1:4000 of 3.9 mg/mL), IBA1 (Wako Pure, Neuss, Germany, 019-19741, 1:1000 of 0.5 mg/mL), and Ki67 (Abcam, Cambridge, UK, #15580, 1:1000 of 0.8 mg/mL). After washing, biotinylated anti-rabbit secondary antibodies were applied, followed by incubation with ExtrAvidin peroxidase. Visualization was achieved using an AEC substrate, and sections were mounted with a gelatin-coating solution. Isotype controls were included using rabbit IgG at equivalent concentrations to the primary antibodies.

4.16. Data Acquisition and Analysis

All slides were scanned using an Axioscan Z1 (Zeiss, Germany) with a 20× objective (EC-Plan Neofluar 20×/0.5) and a color camera (Hitachi HV-F202SCL), yielding a scaling of 0.221 µm × 0.221 µm per pixel.
Quantitative analysis was performed using HALO® 3.3, with a region of interest (ROI) covering the extent of the tumor (KI67, IBA1, and GFAP). Additionally, the stroma was analyzed for IBA1 and GFAP, using a function of HALO® to define a new area (150 µm) around the tumor’s ROI. Empty areas (e.g., from ventricles or defects) and artifacts (blushes or folds) were excluded from the analysis through HALO’s built-in exclusion tool.
The primary analysis for Ki-67 was performed using the CytoNuclear module, with outcome measures being total and Ki-67-positive cells per mm2, along with Ki-67 staining intensity (weak, moderate, and strong). An additional control involved using the Area quantification module to verify segmentation results.
IBA1 and GFAP analysis was performed using the HALO® Microglia module. Here, negative and immune-positive cells per mm2 were the primary outcome measures. As glial morphology within the tumor was changed substantially for IBA1, a secondary analysis (see Supplemental Materials) using the Area quantification module was performed for IBA1 and GFAP staining, confirming findings from the segmentation algorithm.
All quantitative analysis settings were tuned and tested on multiple images. For the subsequent analysis, a single setting was employed for all images. Data from one slide was averaged, and the mean was calculated across the different anatomical planes.

4.17. Statistical Analysis

Data are presented as mean ± SD or SEM from at least three independent experiments. Statistical analyses were performed using GraphPad Prism 9. To compare the effects of U87wt or U87COX−2KO implantation, a two-tailed t-test was used after checking for the normality of the data (Kolmogorov–Smirnov). The relationship between staining intensity and COX-2 phenotype was investigated using a two-way ANOVA with Šídák’s multiple comparisons test, if appropriate. For all analyses, a value of p < 0.05 was considered statistically significant (* p < 0.05, ** p < 0.01, and *** p < 0.001).

5. Conclusions

In conclusion, these findings highlight the potential of complete COX-2 knockout as a therapeutic strategy in GBM, particularly when combined with targeting peripheral macrophages and maintaining a stable microglia population to enhance apoptotic cell clearance and suppress glioblastoma progression. These multi-target therapies, including tissue-specific CRISPR/Cas9 genome editing, hold significant promise in addressing GBM heterogeneity [59].

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijms26136198/s1. References [60,61,62] are cited in the Supplementary Materials.

Author Contributions

C.H.-K., M.T., C.K.D., A.D., S.M., M.L., J.W., J.R., W.D.-C. and J.P. contributed to the acquisition of data. C.H.-K., C.K.D., A.D., M.L., F.H. and J.P. performed the analysis and interpretation of the data. C.H.-K., C.K.D. and J.P. drafted the manuscript. C.H.-K., C.K.D. and J.P. contributed to the study concept and design. C.H.-K. performed the study supervision. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded, in part, by the Deutsche Forschungsgemeinschaft (PI 304/7-1, J.P. and M.L.; Transregio 314061271-CRC/TRR 205/1-2, B10, J.P). Partial support by the LiSyM Cancer phase I joint collaborative project DEEP-HCC (Federal Ministry of Education and Research; No. 031L0258B, J.P. and M.L.) is also acknowledged.

Institutional Review Board Statement

Animal experiments were conducted in accordance with German Animal Welfare Regulations and approved by the local Animal Ethics Committee (Landesdirektion Dresden, Germany) (25-5131/449/39). The study did not involve humans.

Informed Consent Statement

Not applicable.

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Acknowledgments

We appreciate the excellent technical assistance of Aline Morgenegg, Mareike Barth, and Julia Aldinger. The authors thank the head and staff of the animal research facility at HZDR, Birgit Belter, Katrin Baumgart, and Helge Gläser. We thank Torsten Knieß and Martin Kreller from the department of Targetry, Target Chemistry and Radiopharmacy for providing the radiotracers [18F]FDG and [18F]FMISO. The authors also thank Maria Mehnert for performing the tube formation and adhesion assay, which were part of her master’s thesis at the Technische Universität Dresden, Department of Chemistry and Food Chemistry. Furthermore, we would like to thank Jens Stieler and Markus Morawski for scanning the mouse-brain slices on the Axioscan 7 slide scanner at the University of Leipzig “Paul-Flechsig-Institute for Brain Research”, Department of Molecular and Cellular Mechanisms of Neurodegeneration. We would like to thank the National Center for Tumor Diseases (NCT), Partner Site Dresden, for financial support regarding the Brillouin system. Moreover, we would like to thank Ilker Y. Eyüpoglu, director of the Department of Neurosurgery at the University Hospital Carl Gustav Carus Dresden, for access to the Brillouin system.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Mitusova, K.; Peltek, O.O.; Karpov, T.E.; Muslimov, A.R.; Zyuzin, M.V.; Timin, A.S. Overcoming the Blood-Brain Barrier for the Therapy of Malignant Brain Tumor: Current Status and Prospects of Drug Delivery Approaches. J. Nanobiotechnol. 2022, 20, 412. [Google Scholar] [CrossRef] [PubMed]
  2. Ostrom, Q.T.; Price, M.; Neff, C.; Cioffi, G.; Waite, K.A.; Kruchko, C.; Barnholtz-Sloan, J.S. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2015–2019. Neuro Oncol. 2022, 24, v1–v95. [Google Scholar] [CrossRef]
  3. Buonfiglioli, A.; Hambardzumyan, D. Macrophages and Microglia: The Cerberus of Glioblastoma. Acta Neuropathol. Commun. 2021, 9, 54. [Google Scholar] [CrossRef]
  4. Hambardzumyan, D.; Gutmann, D.H.; Kettenmann, H. The Role of Microglia and Macrophages in Glioma Maintenance and Progression. Nat. Neurosci. 2016, 19, 20–27. [Google Scholar] [CrossRef]
  5. Phillips, H.S.; Kharbanda, S.; Chen, R.; Forrest, W.F.; Soriano, R.H.; Wu, T.D.; Misra, A.; Nigro, J.M.; Colman, H.; Soroceanu, L.; et al. Molecular Subclasses of High-Grade Glioma Predict Prognosis, Delineate a Pattern of Disease Progression, and Resemble Stages in Neurogenesis. Cancer Cell 2006, 9, 157–173. [Google Scholar] [CrossRef]
  6. Chen, H.C.; Chang, W.C.; Chuang, J.Y.; Chang, K.Y.; Liou, J.P.; Hsu, T.I. The Complex Role of Eicosanoids in the Brain: Implications for Brain Tumor Development and Therapeutic Opportunities. Biochim. Biophys. Acta Rev. Cancer 2023, 1878, 188957. [Google Scholar] [CrossRef]
  7. Jiang, J.; Qiu, J.; Li, Q.; Shi, Z. Prostaglandin E2 Signaling: Alternative Target for Glioblastoma? Trends Cancer 2017, 3, 75–78. [Google Scholar] [CrossRef]
  8. Qiu, J.; Shi, Z.; Jiang, J. Cyclooxygenase-2 in Glioblastoma Multiforme. Drug Discov. Today 2017, 22, 148–156. [Google Scholar] [CrossRef]
  9. Ferreira, M.T.; Miyake, J.A.; Gomes, R.N.; Feitoza, F.; Stevannato, P.B.; da Cunha, A.S.; Serachi, F.O.; Panagopoulos, A.T.; Colquhoun, A. Cyclooxygenase Inhibition Alters Proliferative, Migratory, and Invasive Properties of Human Glioblastoma Cells In Vitro. Int. J. Mol. Sci. 2021, 22, 4297. [Google Scholar] [CrossRef]
  10. Lombardi, F.; Augello, F.R.; Artone, S.; Ayroldi, E.; Giusti, I.; Dolo, V.; Cifone, M.G.; Cinque, B.; Palumbo, P. Cyclooxygenase-2 Upregulated by Temozolomide in Glioblastoma Cells Is Shuttled In Extracellular Vesicles Modifying Recipient Cell Phenotype. Front. Oncol. 2022, 12, 933746. [Google Scholar] [CrossRef]
  11. Brat, D.J.; Castellano-Sanchez, A.A.; Hunter, S.B.; Pecot, M.; Cohen, C.; Hammond, E.H.; Devi, S.N.; Kaur, B.; Van Meir, E.G. Pseudopalisades in Glioblastoma Are Hypoxic, Express Extracellular Matrix Proteases, and Are Formed by an Actively Migrating Cell Population. Cancer Res. 2004, 64, 920–927. [Google Scholar] [CrossRef] [PubMed]
  12. Kaur, B.; Khwaja, F.W.; Severson, E.A.; Matheny, S.L.; Brat, D.J.; Van Meir, E.G. Hypoxia and the Hypoxia-Inducible-Factor Pathway in Glioma Growth and Angiogenesis. Neuro Oncol. 2005, 7, 134–153. [Google Scholar] [CrossRef] [PubMed]
  13. Park, J.H.; Lee, H.K. Current Understanding of Hypoxia in Glioblastoma Multiforme and Its Response to Immunotherapy. Cancers 2022, 14, 1176. [Google Scholar] [CrossRef] [PubMed]
  14. Qiu, Y.; Li, P.; Ji, C. Cell Death Conversion under Hypoxic Condition in Tumor Development and Therapy. Int. J. Mol. Sci. 2015, 16, 25536–25551. [Google Scholar] [CrossRef]
  15. Rodrigues, P.; Bangali, H.; Hammoud, A.; Mustafa, Y.F.; Al-Hetty, H.R.A.K.; Alkhafaji, A.T.; Deorari, M.M.; Al-Taee, M.M.; Zabibah, R.S.; Alsalamy, A. COX 2-Inhibitors; a Thorough and Updated Survey into Combinational Therapies in Cancers. Med. Oncol. 2024, 41, 41. [Google Scholar] [CrossRef]
  16. Haase-Kohn, C.; Wolf, S.; Herwig, N.; Mosch, B.; Pietzsch, J. Metastatic Potential of B16-F10 Melanoma Cells Is Enhanced by Extracellular S100A4 Derived from RAW264.7 Macrophages. Biochem. Biophys. Res. Commun. 2014, 446, 143–148. [Google Scholar] [CrossRef]
  17. Dean, P.T.; Hooks, S.B. Pleiotropic Effects of the COX-2/PGE2 Axis in the Glioblastoma Tumor Microenvironment. Front. Oncol. 2022, 12, 1116014. [Google Scholar] [CrossRef]
  18. Lombardi, F.; Augello, F.R.; Artone, S.; Ciafarone, A.; Topi, S.; Cifone, M.G.; Cinque, B.; Palumbo, P. Involvement of Cyclooxygenase-2 in Establishing an Immunosuppressive Microenvironment in Tumorspheres Derived from TMZ-Resistant Glioblastoma Cell Lines and Primary Cultures. Cells 2024, 13, 258. [Google Scholar] [CrossRef]
  19. Knudsen, A.M.; Ewald, J.D.; Pedersen, V.; Haupt-Jorgensen, M.; Hansen, E.V.R.; Kristensen, B.W. Characterisation of the tumour microenvironment in primary and recurrent glioblastomas. Neuropathol Appl Neurobiol. 2024, 50, 5. [Google Scholar] [CrossRef] [PubMed]
  20. Han, Y.; Liu, D.; Li, L. PD-1/PD-L1 Pathway: Current Researches in Cancer. Am. J. Cancer Res. 2020, 10, 727–742. [Google Scholar]
  21. Radin, D.P.; Tsirka, S.E. Interactions between Tumor Cells, Neurons, and Microglia in the Glioma Microenvironment. Int. J. Mol. Sci. 2020, 21, 8476. [Google Scholar] [CrossRef] [PubMed]
  22. Hashemi Goradel, N.; Najafi, M.; Salehi, E.; Farhood, B.; Mortezaee, K. Cyclooxygenase-2 in Cancer: A Review. J. Cell Physiol. 2019, 234, 5683–5699. [Google Scholar] [CrossRef] [PubMed]
  23. Frederico, S.C.; Zhang, X.; Hu, B.; Kohanbash, G. Pre-Clinical Models for Evaluating Glioma Targeted Immunotherapies. Front. Immunol. 2022, 13, 1092399. [Google Scholar] [CrossRef] [PubMed]
  24. Gao, L.; Wang, T.H.; Chen, C.P.; Xiang, J.J.; Zhao, X.-B.; Gui, R.-Y.; Liao, X.-H. Targeting COX-2 Potently Inhibits Proliferation of Cancer Cells in Vivo but Not in Vitro in Cutaneous Squamous Cell Carcinoma. Transl. Cancer Res. 2021, 10, 2219–2228. [Google Scholar] [CrossRef]
  25. Lv, X.; Chen, Z.; Li, S.; Xie, H. Knockdown of Cyclooxygenase-2 Leads to Growth Inhibition and Cell Cycle Arrest in Hepatocellular Carcinoma Cells. Onco Targets Ther. 2019, 12, 4341–4349. [Google Scholar] [CrossRef]
  26. Haase-Kohn, C.; Laube, M.; Donat, C.K.; Belter, B.; Pietzsch, J. CRISPR/Cas9 Mediated Knockout of Cyclooxygenase-2 Gene Inhibits Invasiveness in A2058 Melanoma Cells. Cells 2022, 11, 749. [Google Scholar] [CrossRef]
  27. Kniess, T.; Laube, M.; Bergmann, R.; Sehn, F.; Graf, F.; Steinbach, J.; Wuest, F.; Pietzsch, J. Radiosynthesis of a ¹⁸F-labeled 2,3-diarylsubstituted indole via McMurry coupling for functional characterization of cyclooxygenase-2 (COX-2) in vitro and in vivo. Bio. Med. Chem. 2012, 11, 3410–3421. [Google Scholar] [CrossRef]
  28. Ullrich, M.; Richter, S.; Seifert, V.; Hauser, S.; Calsina, B.; Martínez-Montes, A.M.; Ter Laak, M.; Ziegler, C.; Timmers, H.; Eisenhofer, G.; et al. Targeting Cyclooxygenase-2 in Pheochromocytoma and Paraganglioma: Focus on Genetic Background. Cancer 2019, 11, 743. [Google Scholar] [CrossRef]
  29. Lombardi, F.; Augello, F.R.; Artone, S.; Gugu, M.K.; Cifone, M.G.; Cinque, B.; Palumbo, P. Up-Regulation of Cyclooxygenase-2 (COX-2) Expression by Temozolomide (TMZ) in Human Glioblastoma (GBM) Cell Lines. Int. J. Mol. Sci. 2022, 23, 1545. [Google Scholar] [CrossRef]
  30. Palumbo, P.; Lombardi, F.; Augello, F.R.; Giusti, I.; Dolo, V.; Leocata, P.; Cifone, M.G.; Cinque, B. Biological Effects of Selective COX-2 Inhibitor NS398 on Human Glioblastoma Cell Lines. Cancer Cell Int. 2020, 20, 167. [Google Scholar] [CrossRef]
  31. Yu, T.; Lao, X.; Zheng, H. Influencing COX-2 Activity by COX Related Pathways in Inflammation and Cancer. Mini Rev. Med. Chem. 2016, 16, 1230–1243. [Google Scholar] [CrossRef] [PubMed]
  32. Wu, Z.; Nakanishi, H. Phosphatidylserine-Containing Liposomes: Potential Pharmacological Interventions against Inflammatory and Immune Diseases through the Production of Prostaglandin E(2) after Uptake by Myeloid Derived Phagocytes. Arch. Immunol. Et Ther. Exp. 2011, 59, 195–201. [Google Scholar] [CrossRef] [PubMed]
  33. Kulesza, A.; Paczek, L.; Burdzinska, A. The Role of COX-2 and PGE2 in the Regulation of Immunomodulation and Other Functions of Mesenchymal Stromal Cells. Biomedicines 2023, 11, 445. [Google Scholar] [CrossRef] [PubMed]
  34. Pavlacky, J.; Polak, J. Technical Feasibility and Physiological Relevance of Hypoxic Cell Culture Models. Front. Endocrinol. 2020, 11, 57. [Google Scholar] [CrossRef]
  35. Keeley, T.P.; Mann, G.E. Defining Physiological Normoxia for Improved Translation of Cell Physiology to Animal Models and Humans. Physiol. Rev. 2019, 99, 161–234. [Google Scholar] [CrossRef]
  36. Rana, R.; Huirem, R.S.; Kant, R.; Chauhan, K.; Sharma, S.; Yashavarddhan, M.H.; Chhabra, S.S.; Acharya, R.; Kalra, S.K.; Gupta, A.; et al. Cytochrome C as a Potential Clinical Marker for Diagnosis and Treatment of Glioma. Front. Oncol. 2022, 12, 960787. [Google Scholar] [CrossRef]
  37. Kalpage, H.A.; Bazylianska, V.; Recanati, M.A.; Fite, A.; Liu, J.; Wan, J.; Mantena, N.; Malek, M.H.; Podgorski, I.; Heath, E.I.; et al. Tissue-Specific Regulation of Cytochrome c by Post-Translational Modifications: Respiration, the Mitochondrial Membrane Potential, ROS, and Apoptosis. FASEB J. 2019, 33, 1540–1553. [Google Scholar] [CrossRef]
  38. Ott, M.; Robertson, J.D.; Gogvadze, V.; Zhivotovsky, B.; Orrenius, S. Cytochrome c Release from Mitochondria Proceeds by a Two-Step Process. Proc. Natl. Acad. Sci. USA 2002, 99, 1259–1263. [Google Scholar] [CrossRef]
  39. Colecchia, D.; Strambi, A.; Sanzone, S.; Iavarone, C.; Rossi, M.; Dall’Armi, C.; Piccioni, F.; Verrotti di Pianella, A.; Chiariello, M. MAPK15/ERK8 Stimulates Autophagy by Interacting with LC3 and GABARAP Proteins. Autophagy 2012, 8, 1724–1740. [Google Scholar] [CrossRef]
  40. Dutta, S.; Mahalanobish, S.; Saha, S.; Mandal, M.; Begam, S.; Sadhukhan, P.; Ghosh, S.; Brahmachari, G.; Sil, P.C. Biological Evaluation of the Novel 3,3′-((4-Nitrophenyl)Methylene)Bis(4-Hydroxy-2H-Chromen-2-One) Derivative as Potential Anticancer Agents via the Selective Induction of Reactive Oxygen Species-Mediated Apoptosis. Cell Signal 2023, 111, 110876. [Google Scholar] [CrossRef]
  41. Kern, M.A.; Haugg, A.M.; Koch, A.F.; Schilling, T.; Breuhahn, K.; Walczak, H.; Fleischer, B.; Trautwein, C.; Michalski, C.; Schulze-Bergkamen, H.; et al. Cyclooxygenase-2 Inhibition Induces Apoptosis Signaling via Death Receptors and Mitochondria in Hepatocellular Carcinoma. Cancer Res. 2006, 66, 7059–7066. [Google Scholar] [CrossRef] [PubMed]
  42. Sha, W.; Olesch, C.; Hanaka, H.; Rådmark, O.; Weigert, A.; Brüne, B. Necrosis in DU145 Prostate Cancer Spheroids Induces COX-2/mPGES-1-Derived PGE2 to Promote Tumor Growth and to Inhibit T Cell Activation. Int. J. Cancer 2013, 133, 1578–1588. [Google Scholar] [CrossRef] [PubMed]
  43. Wang, G.; Zhong, K.; Wang, Z.; Zhang, Z.; Tang, X.; Tong, A.; Zhou, L. Tumor-Associated Microglia and Macrophages in Glioblastoma: From Basic Insights to Therapeutic Opportunities. Front. Immunol. 2022, 13, 964898. [Google Scholar] [CrossRef]
  44. Acker, G.; Zollfrank, J.; Jelgersma, C.; Nieminen-Kelha, M.; Kremenetskaia, I.; Mueller, S.; Ghori, A.; Vajkoczy, P.; Brandenburg, S. The CXCR2/CXCL2 Signalling Pathway—An Alternative Therapeutic Approach in High-Grade Glioma. Eur. J. Cancer 2020, 126, 106–115. [Google Scholar] [CrossRef]
  45. Brandenburg, S.; Müller, A.; Turkowski, K.; Radev, Y.T.; Rot, S.; Schmidt, C.; Bungert, A.D.; Acker, G.; Schorr, A.; Hippe, A.; et al. Resident Microglia Rather than Peripheral Macrophages Promote Vascularization in Brain Tumors and Are Source of Alternative Pro-Angiogenic Factors. Acta Neuropathol. 2016, 131, 365–378. [Google Scholar] [CrossRef]
  46. Ercolano, G.; De Cicco, P.; Rubino, V.; Terrazzano, G.; Ruggiero, G.; Carriero, R.; Kunderfranco, P.; Ianaro, A. Knockdown of PTGS2 by CRISPR/CAS9 System Designates a New Potential Gene Target for Melanoma Treatment. Front. Pharmacol. 2019, 10, 1456. [Google Scholar] [CrossRef]
  47. Leite, D.M.; Zvar Baskovic, B.; Civita, P.; Neto, C.; Gumbleton, M.; Pilkington, G.J. A Human Co-Culture Cell Model Incorporating Microglia Supports Glioblastoma Growth and Migration, and Confers Resistance to Cytotoxics. FASEB J. 2020, 34, 1710–1727. [Google Scholar] [CrossRef]
  48. Kolliopoulos, C.; Ali, M.M.; Castillejo-Lopez, C.; Heldin, C.H.; Heldin, P. CD44 Depletion in Glioblastoma Cells Suppresses Growth and Stemness and Induces Senescence. Cancers 2022, 14, 3747. [Google Scholar] [CrossRef]
  49. Dugandzija, T.; Drljaca, J.; Bulajic, D.; Isakovic, A.; Stilinovic, N.; Sekulic, S.; Capo, I. Hallmarks of Tumor-Associated Microglia Response to Experimental U87 Human Glioblastoma Xenograft. Tissue Cell 2021, 72, 101557. [Google Scholar] [CrossRef]
  50. Simone, L.; Capobianco, D.L.; Di Palma, F.; Binda, E.; Legnani, F.G.; Vescovi, A.L.; Svelto, M.; Pisani, F. GFAP Serves as a Structural Element of Tunneling Nanotubes between Glioblastoma Cells and Could Play a Role in the Intercellular Transfer of Mitochondria. Front. Cell Dev. Biol. 2023, 11, 1221671. [Google Scholar] [CrossRef]
  51. Tang, G.; Yue, Z.; Talloczy, Z.; Hagemann, T.; Cho, W.; Messing, A.; Sulzer, D.L.; Goldman, J.E. Autophagy Induced by Alexander Disease-Mutant GFAP Accumulation Is Regulated by P38/MAPK and mTOR Signaling Pathways. Hum. Mol. Genet. 2008, 17, 1540–1555. [Google Scholar] [CrossRef] [PubMed]
  52. Rong, X.; Huang, B.; Qiu, S.; Li, X.; He, L.; Peng, Y. Tumor-Associated Macrophages Induce Vasculogenic Mimicry of Glioblastoma Multiforme through Cyclooxygenase-2 Activation. Oncotarget 2016, 7, 83976–83986. [Google Scholar] [CrossRef] [PubMed]
  53. Tjiu, J.W.; Chen, J.S.; Shun, C.T.; Lin, S.J.; Liao, Y.H.; Chu, C.Y.; Tsai, T.F.; Chiu, H.C.; Dai, Y.S.; Inoue, H.; et al. Tumor-Associated Macrophage-Induced Invasion and Angiogenesis of Human Basal Cell Carcinoma Cells by Cyclooxygenase-2 Induction. J. Investig. Dermatol. 2009, 129, 1016–1025. [Google Scholar] [CrossRef]
  54. Zheng, T.; Chen, K.; Zhang, X.; Feng, H.; Shi, Y.; Liu, L.; Zhang, J.; Chen, Y. Knockdown of TXNDC9 Induces Apoptosis and Autophagy in Glioma and Mediates Cell Differentiation by P53 Activation. Aging 2020, 12, 18649–18659. [Google Scholar] [CrossRef]
  55. Paxinos, G.; Franklin, K.B.J. The Mouse Brain in Stereotaxic Coordinates, 2nd ed.; Academic Press: Cambridge, MA, USA, 2021. [Google Scholar]
  56. Erö, C.; Gewaltig, M.O.; Keller, D.; Markram, H. A Cell Atlas for the Mouse Brain. Front. Neuroinform. 2019, 28, 13. [Google Scholar] [CrossRef]
  57. Rodarie, D.; Verasztó, C.; Roussel, J.; Reimann, M.; Keller, D.; Ramaswamy, S.; Markram, H.; Gewaltig, M.O. A method to estimate the cellular composition of the mouse brain from heterogeneous datasets. PLoS Comput. Biol. 2022, 18, 12. [Google Scholar] [CrossRef]
  58. Roussel, Y.; Verasztó, C.; Rodarie, D.; Damart, T.; Reimann, M.; Ramaswamy, S.; Markram, H.; Keller, D. Mapping of morpho-electric features to molecular identity of cortical inhibitory neurons. PLoS Comput. Biol. 2023, 19, e1010058. [Google Scholar] [CrossRef]
  59. Rabaan, A.A.; AlSaihati, H.; Bukhamsin, R.; Bakhrebah, M.A.; Nassar, M.S.; Alsaleh, A.A.; Alhashem, Y.N.; Bukhamseen, A.Y.; Al-Ruhimy, K.; Alotaibi, M.; et al. Application of CRISPR/Cas9 Technology in Cancer Treatment: A Future Direction. Curr. Oncol. 2023, 30, 1954–1976. [Google Scholar] [CrossRef]
  60. Mosch, B.; Mueller, K.; Steinbach, J.; Pietzsch, J. Influence of irradiation on metabolism and metastatic potential of B16-F10 melanoma cells. Int. J. Radiat. Biol. 2009, 85, 1002–1012. [Google Scholar] [CrossRef]
  61. Chen, X.; Zheng, Q.; Li, W.; Lu, Y.; Ni, Y.; Ma, L.; Fu, Y. SOX5 induces lung adenocarcinoma angiogenesis by inducing the expression of VEGF through STAT3 signaling. Onco Targets Ther. 2018, 11, 5733–5741. [Google Scholar] [CrossRef]
  62. Rix, J.; Uckermann, O.; Kirsche, K.; Schackert, G.; Koch, E.; Kirsch, M.; Galli, R. Correlation of biomechanics and cancer cell phenotype by combined Brillouin and Raman spectroscopy of U87-MG glioblastoma cells. J. R. Soc. Interface 2022, 19, 209. [Google Scholar] [CrossRef]
Ijms 26 06198 g001
Figure 1. A comparative analysis of COX-2 knockout (red) and wildtype (black) U87 monolayer cell cultures. (a) Analysis of cellular growth of U87wt and U87COX−2KO over 9 days (n = 3). (b) Cell cycle analysis under normoxic and hypoxic conditions using flow cytometry after staining with propidium iodide. (c) One sample analysis with six replicates of a “Signaling Phosphor Array” and a “Cancer Biomarker Antibody Array”. Mean values and statistical analysis was provided by the manufacturer. Changes were considered significant if the expression ratio between U87wt and U87COX-2KO was ≤0.5 (decrease) or ≥2.0 (increase). (d) Immunoblot analysis using indicated antibodies. Western blots were performed either in a duplicate (upper panels) or a multiple (lower panels) and normalized to GAPDH. (e) Determination of mitochondria (MitoTracker green via flow cytometry), ROS (CellROX via flow cytometry), and lipid peroxidation (MDA via TBARS Assay) after different stressor treatment for 24h: H2O2 (200 μM) as a model of an oxidative stressor; CoCl2 (200 nM) as a hypoxia-inducing agent; supernatants of HMC3 microglia cells (S_HMC3); media without serum (woFCS); media without Glucose (woGlucose). (f) Uptake of [18F]FDG over 120 min, and [18F]FMISO over 240 min (upper panel) in U87wt (black) and U87COX-2KO (red) cells. Additional uptake of [18F]FMISO after coculture with MG cells (1:1) and different treatments for 24 h. Results of three independent experiments with quadruple determination. Values in the graphs represent the mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001. For (d) * p < 0.05, ** p < 0.01 vs. U87wt. For (f) ** p < 0.01, *** p < 0.001 vs. control; # p < 0.05, ### p < 0.001 vs. treatment; § p < 0.05, §§ p < 0.01 U87wt + MG vs. U87COX-2KO + MG.
Figure 1. A comparative analysis of COX-2 knockout (red) and wildtype (black) U87 monolayer cell cultures. (a) Analysis of cellular growth of U87wt and U87COX−2KO over 9 days (n = 3). (b) Cell cycle analysis under normoxic and hypoxic conditions using flow cytometry after staining with propidium iodide. (c) One sample analysis with six replicates of a “Signaling Phosphor Array” and a “Cancer Biomarker Antibody Array”. Mean values and statistical analysis was provided by the manufacturer. Changes were considered significant if the expression ratio between U87wt and U87COX-2KO was ≤0.5 (decrease) or ≥2.0 (increase). (d) Immunoblot analysis using indicated antibodies. Western blots were performed either in a duplicate (upper panels) or a multiple (lower panels) and normalized to GAPDH. (e) Determination of mitochondria (MitoTracker green via flow cytometry), ROS (CellROX via flow cytometry), and lipid peroxidation (MDA via TBARS Assay) after different stressor treatment for 24h: H2O2 (200 μM) as a model of an oxidative stressor; CoCl2 (200 nM) as a hypoxia-inducing agent; supernatants of HMC3 microglia cells (S_HMC3); media without serum (woFCS); media without Glucose (woGlucose). (f) Uptake of [18F]FDG over 120 min, and [18F]FMISO over 240 min (upper panel) in U87wt (black) and U87COX-2KO (red) cells. Additional uptake of [18F]FMISO after coculture with MG cells (1:1) and different treatments for 24 h. Results of three independent experiments with quadruple determination. Values in the graphs represent the mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001. For (d) * p < 0.05, ** p < 0.01 vs. U87wt. For (f) ** p < 0.01, *** p < 0.001 vs. control; # p < 0.05, ### p < 0.001 vs. treatment; § p < 0.05, §§ p < 0.01 U87wt + MG vs. U87COX-2KO + MG.
Ijms 26 06198 g001
Ijms 26 06198 g002
Figure 2. Comparative analysis of COX-2 knockout (red) to wildtype (black) U87 spheroids. (a) Growth curves of U87wt and U87COX−2KO over 21 days (n = 12). (b) Representative images of spheroids after days 5, 12, and 19 with staining for H&E, pimonidazole, and Ki-67. (c) Uptake of [18F]FDG over 120 min and [18F]FMISO over 240 min in %ID/g. (d) Determination of ROS activity (CellROX) in U87wt and U87COX−2KO spheroids after 5, 12, and 19 days. Results of two independent experiments with quadruple determination. (e) Immunostaining of tissue sections from U87wt and U87COX-2KO spheroids, using indicated antibodies. Values in the graphs (a,c,d) represent the mean ± SEM. *** p < 0.001. For (d) ** p < 0.01, *** p < 0.001 vs. days; ### p < 0.001 U87wt vs. U87COX-2KO. Scale bar: 100 µm.
Figure 2. Comparative analysis of COX-2 knockout (red) to wildtype (black) U87 spheroids. (a) Growth curves of U87wt and U87COX−2KO over 21 days (n = 12). (b) Representative images of spheroids after days 5, 12, and 19 with staining for H&E, pimonidazole, and Ki-67. (c) Uptake of [18F]FDG over 120 min and [18F]FMISO over 240 min in %ID/g. (d) Determination of ROS activity (CellROX) in U87wt and U87COX−2KO spheroids after 5, 12, and 19 days. Results of two independent experiments with quadruple determination. (e) Immunostaining of tissue sections from U87wt and U87COX-2KO spheroids, using indicated antibodies. Values in the graphs (a,c,d) represent the mean ± SEM. *** p < 0.001. For (d) ** p < 0.01, *** p < 0.001 vs. days; ### p < 0.001 U87wt vs. U87COX-2KO. Scale bar: 100 µm.
Ijms 26 06198 g002
Ijms 26 06198 g003
Figure 3. Subcutaneous xenograft growth of U87wt (n = 44) and U87COX−2KO (n = 38) in NMRI nude mice. (a) Comparison of tumor volume over time between U87wt (black) and U87COX−2KO (red). (b) Comparison of tumor volume over time between U87wt (black) with either U87wt + Mϕs (dark green, n = 23) or U87wt + MCs (green, n = 18) (left panel). Comparison of tumor volume over time between U87COX−2KO (red) with either U87COX−2KO + Mϕs (dark green, n = 23) or U87COX−2KO + MCs (green, n = 18) (right panel). (c) Comparison of tumor volume over time between U87wt (black) and U87wt + MG (blue, n = 13) (left panel). Comparison of tumor volume over time between U87COX−2KO (red) and U87COX−2KO + MG (blue, n = 23) (right panel). (d) Radiotracer uptake 240 min after i.v. injection of 8–12 MBq [18F]FMISO in xenografts from U87wt and U87COX−2KO cells or as multi-cellular tumors with MG cells (left panel). Radiotracer uptake 60 min after i.v. injection of 10 MBq [18F]FDG in U87wt and U87COX−2KO (right panel), respectively, in NMRI nude mice xenografts. (SUV: standardized uptake value.) Values in the graphs represent the mean ± SEM. * p < 0.05, ** p < 0.01, *** p <0.001.
Figure 3. Subcutaneous xenograft growth of U87wt (n = 44) and U87COX−2KO (n = 38) in NMRI nude mice. (a) Comparison of tumor volume over time between U87wt (black) and U87COX−2KO (red). (b) Comparison of tumor volume over time between U87wt (black) with either U87wt + Mϕs (dark green, n = 23) or U87wt + MCs (green, n = 18) (left panel). Comparison of tumor volume over time between U87COX−2KO (red) with either U87COX−2KO + Mϕs (dark green, n = 23) or U87COX−2KO + MCs (green, n = 18) (right panel). (c) Comparison of tumor volume over time between U87wt (black) and U87wt + MG (blue, n = 13) (left panel). Comparison of tumor volume over time between U87COX−2KO (red) and U87COX−2KO + MG (blue, n = 23) (right panel). (d) Radiotracer uptake 240 min after i.v. injection of 8–12 MBq [18F]FMISO in xenografts from U87wt and U87COX−2KO cells or as multi-cellular tumors with MG cells (left panel). Radiotracer uptake 60 min after i.v. injection of 10 MBq [18F]FDG in U87wt and U87COX−2KO (right panel), respectively, in NMRI nude mice xenografts. (SUV: standardized uptake value.) Values in the graphs represent the mean ± SEM. * p < 0.05, ** p < 0.01, *** p <0.001.
Ijms 26 06198 g003
Ijms 26 06198 g004
Figure 4. Orthotopic xenograft growth of U87wt and U87COX−2KO in NMRI nude mice. In vivo T2-weighted MRI images and calculated tumor volumes at various timepoints of (a) U87wt implanted tumor cells in 8 out of 8 animals (dotted lines) and (b) U87COX−2KO implanted tumor cells in 5 out of 8 animals (dotted lines), along with corresponding representative images. Value in the graph represents the mean ± SEM. * p < 0.05.
Figure 4. Orthotopic xenograft growth of U87wt and U87COX−2KO in NMRI nude mice. In vivo T2-weighted MRI images and calculated tumor volumes at various timepoints of (a) U87wt implanted tumor cells in 8 out of 8 animals (dotted lines) and (b) U87COX−2KO implanted tumor cells in 5 out of 8 animals (dotted lines), along with corresponding representative images. Value in the graph represents the mean ± SEM. * p < 0.05.
Ijms 26 06198 g004
Ijms 26 06198 g005
Figure 5. The qualitative (H&E) and quantitative histology of Ki-67, IBA1-, and GFAP-positive cells in the tumor. Representative whole brain photomicrographs for H&E (a), Ki-67 (b), IBA1 (c), and GFAP (d) staining are presented in the left panel (upper micrograph for U87wt and lower for U87COX-KO 2KO, respectively). A black rectangle indicates the magnified area shown in the middle left panel. A further magnification (yellow rectangle) and its location are provided in the middle right panel. For H&E, the right panel shows a second (blue rectangle) and a third magnification (yellow rectangle). For all other stains, the right panel shows the segmentation results as a color-coded overlay. For Ki-67 (B, right panel), this indicates hematoxylin-positive (blue) and Ki-67-positive cells, color-coded for staining intensity (yellow: weak; orange: moderate; and red: strong). For IBA1 and GFAP (C/D, right panel), this indicates hematoxylin-positive (blue) and IBA1/GFAP-positive cells (green). The processes are labeled yellow (quantified but not analyzed). Graphs show the mean ± SD of stain-positive cells per animal (from 6–12 sections) of U87wt and U87COX2-KO (n = 5 each). Scale bars are color-coded with light blue: 1000 µm (whole brain photomicrographs); solid black: 500 µm; red: 100 µm and white: 50 µm (all magnified views). * p < 0.05, ** p < 0.01 U87wt vs. U87COX2-KO.
Figure 5. The qualitative (H&E) and quantitative histology of Ki-67, IBA1-, and GFAP-positive cells in the tumor. Representative whole brain photomicrographs for H&E (a), Ki-67 (b), IBA1 (c), and GFAP (d) staining are presented in the left panel (upper micrograph for U87wt and lower for U87COX-KO 2KO, respectively). A black rectangle indicates the magnified area shown in the middle left panel. A further magnification (yellow rectangle) and its location are provided in the middle right panel. For H&E, the right panel shows a second (blue rectangle) and a third magnification (yellow rectangle). For all other stains, the right panel shows the segmentation results as a color-coded overlay. For Ki-67 (B, right panel), this indicates hematoxylin-positive (blue) and Ki-67-positive cells, color-coded for staining intensity (yellow: weak; orange: moderate; and red: strong). For IBA1 and GFAP (C/D, right panel), this indicates hematoxylin-positive (blue) and IBA1/GFAP-positive cells (green). The processes are labeled yellow (quantified but not analyzed). Graphs show the mean ± SD of stain-positive cells per animal (from 6–12 sections) of U87wt and U87COX2-KO (n = 5 each). Scale bars are color-coded with light blue: 1000 µm (whole brain photomicrographs); solid black: 500 µm; red: 100 µm and white: 50 µm (all magnified views). * p < 0.05, ** p < 0.01 U87wt vs. U87COX2-KO.
Ijms 26 06198 g005
Ijms 26 06198 g006
Figure 6. Overview of experimental details and protocol for in vivo experiments and downstream quantitative histology. (a) Brain region chosen for injection with coordinates and schematic of approximate tumor size. (b) Experimental pipeline for animal experiments. (c) Tissue processing, sectioning, and immunostaining procedures. (d) Quantitative histology workflow from imaging, region of interest creation, and analysis type to outcome measures used for statistical comparison. Figure a was created using material from [55] and the “Blue Brain Cell Atlas” (URL accessed 12/11/2024) based on publications by [56,57,58].
Figure 6. Overview of experimental details and protocol for in vivo experiments and downstream quantitative histology. (a) Brain region chosen for injection with coordinates and schematic of approximate tumor size. (b) Experimental pipeline for animal experiments. (c) Tissue processing, sectioning, and immunostaining procedures. (d) Quantitative histology workflow from imaging, region of interest creation, and analysis type to outcome measures used for statistical comparison. Figure a was created using material from [55] and the “Blue Brain Cell Atlas” (URL accessed 12/11/2024) based on publications by [56,57,58].
Ijms 26 06198 g006
Table 1. Relative growth rate for xenografts from U87wt and U87COX-2KO cells and multi-cellular tumors with either Mϕ, MC, or MG.
Table 1. Relative growth rate for xenografts from U87wt and U87COX-2KO cells and multi-cellular tumors with either Mϕ, MC, or MG.
NMean ± SDMedian
U87wt442.0 ± 1.01.8
U87wt + MC182.3 ± 1.11.7
U87wt + Mϕ232.1 ± 1.11.8
U87wt + MG133.1 ± 0.43.0
U87COX-2KO380.9 ± 0.41.0
U87COX-2KO+MC182.0 ± 0.71.8
U87COX-2KO +Mϕ231.8 ± 0.71.8
U87COX-2KO + MG230.4 ± 0.60.0
Table 2. Relative growth rate along with statistics for xenografts from U87wt and U87COX-2KO cells and multi-cellular tumors with either Mϕ, MC, or MG.
Table 2. Relative growth rate along with statistics for xenografts from U87wt and U87COX-2KO cells and multi-cellular tumors with either Mϕ, MC, or MG.
Parametersp Value
U87wt vs U87COX-2KO<0.001
U87wt vs U87wt + MC0.57
U87wt vs U87wt + Mϕ0.98
U87wt vs U87wt + MG0.3
U87COX-2KO vs U87COX-2KO + MC<0.001
U87COX-2KO vs U87COX-2KO + Mϕ<0.001
U87COX-2KO vs U87COX-2KO + MG<0.002
U87wt + MC vs U87COX-2KO + MC0.78
U87wt + Mϕ vs U87COX-2KO + Mϕ0.67
U87wt + MG vs U87COX-2KO + MG<0.001
Analyses were performed with two-sided Wilcoxon test.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Pietzsch, J.; Toussaint, M.; Donat, C.K.; Doctor, A.; Meister, S.; Wodtke, J.; Laube, M.; Hofheinz, F.; Rix, J.; Deuther-Conrad, W.; et al. Exploring the Role of Peripheral Macrophages in Glioma Progression: The Metabolic Significance of Cyclooxygenase-2 (COX-2). Int. J. Mol. Sci. 2025, 26, 6198. https://doi.org/10.3390/ijms26136198

AMA Style

Pietzsch J, Toussaint M, Donat CK, Doctor A, Meister S, Wodtke J, Laube M, Hofheinz F, Rix J, Deuther-Conrad W, et al. Exploring the Role of Peripheral Macrophages in Glioma Progression: The Metabolic Significance of Cyclooxygenase-2 (COX-2). International Journal of Molecular Sciences. 2025; 26(13):6198. https://doi.org/10.3390/ijms26136198

Chicago/Turabian Style

Pietzsch, Jens, Magali Toussaint, Cornelius Kurt Donat, Alina Doctor, Sebastian Meister, Johanna Wodtke, Markus Laube, Frank Hofheinz, Jan Rix, Winnie Deuther-Conrad, and et al. 2025. "Exploring the Role of Peripheral Macrophages in Glioma Progression: The Metabolic Significance of Cyclooxygenase-2 (COX-2)" International Journal of Molecular Sciences 26, no. 13: 6198. https://doi.org/10.3390/ijms26136198

APA Style

Pietzsch, J., Toussaint, M., Donat, C. K., Doctor, A., Meister, S., Wodtke, J., Laube, M., Hofheinz, F., Rix, J., Deuther-Conrad, W., & Haase-Kohn, C. (2025). Exploring the Role of Peripheral Macrophages in Glioma Progression: The Metabolic Significance of Cyclooxygenase-2 (COX-2). International Journal of Molecular Sciences, 26(13), 6198. https://doi.org/10.3390/ijms26136198

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop